Acoustic 150 Output Section

[gbono]

Hello,

   First time poster - very interesting site.

  I have an Acoustic 150 which develops a loss of output and starts to distort after the amp warms up. The problem tracks both pre-amp boards so I believe that the issue is with the output board. I have replaced the output coupling capacitor, checked the thermistors, bias resistors and still have the issue. Any help appreciated.


   Also, why did many of the designs of this vintage use a transformer on the input of the output transistors? Was this for impedance matching?

THX
gb


   

     

[joecool85]

I don't know anything about this particular amp, but it sounds to me like it's over heating.  Check and make sure none of the vents are blocked off, and also check and make sure the heatsink(s) are attatched firmly.  It probably wouldn't hurt to pull the chips off the heatsinks, and then re-apply thermal paste and put it back together.  If you do that, clean everything up real nice before putting on the new paste.

[gbono]

Thanks for the comments - with the chassis on the bench- the transistors can still be touched (50C) when the amp is exhibiting issues. I'll try and re-seat the output transistor but I think the problem is somewhere in the drive circuitry.

gb

[teemuk]

"Also, why did many of the designs of this vintage use a transformer on the input of the output transistors? Was this for impedance matching?"

Not really. Those transformer coupled designs practically just follow the old topologies of tube amps that did not use an output transformer. For example, this one: http://web.comhem.se/illuwatar/Tube/Headphone/images/amp_schematic.gif

Back in the days manufacturing decent PNP power transistors was quite difficult and they were either unobtainable or very expensive. As you can notice, amps using the interstage transformer (and later quasi-complementary output) have the same polarity output devices (usually NPN). The interstage transformer acts as signal inverter for the other output transistor - practically converting NPN to PNP. If you know how quasi-complementary output stage (which combines Sziklai and Darlington pairs) looks like (and works) you can visualize this configuration better.

Basically, the transformer is an impedance matching device (which, converts the secondary side impedance to a suitable collector - or emitter - load) ...but that just an inherent property of transformers in general. The transformer is not needed for impedance matching but signal inversion. If you take a look at basic QC topology (ie. http://sound.westhost.com/p12a-f1.gif) you can see that it uses a PNP low-power transistor in common emitter configuration, which results in inverted signal appearing at its collector. So, the voltage amplifier stage can be directly coupled to emitter follower current amplifier stage without impedance matching as well.

Once the transformer is eliminated from the signal path, the phase shifts introduced by it are eliminated as well and the total amount of negative feedback can be raised higher. This in turn results into increase in bandwidth and decrease in output impedance, which in turn means a more linear amplifier. Also, distortions and other less-ideal phenomenons introduced by the transformer (i.e. bandwidth limit, saturation) are eliminated. Very nice for HiFi but not neccessarily so for guitar amps. Anyway, the latter is a subjective issue and depends hugely on the circuits.

I hope this answer satisfied your curiosity.


Edit:

And speaking about that distortion issue: Does the distortion sound like crossover distortion (annoying buzz on low volume signals/"metallic" highs) or like clipping into rails ("fuzz")? This should (likely) help you to pinpoint the problem either to bias/thermal tracking issues or to voltage amplifier stage that is transformer coupled to output buffer.



In case someone didn't know, schematics can be found here: http://acoustic360.homeunix.net/schematics.html